{"id":19467579,"url":"https://github.com/thunlp/ikrl","last_synced_at":"2025-04-25T11:31:25.659Z","repository":{"id":89615970,"uuid":"92007310","full_name":"thunlp/IKRL","owner":"thunlp","description":"Image-embodied Knowledge Representation Learning (IJCAI-2017)","archived":false,"fork":false,"pushed_at":"2021-11-16T04:27:33.000Z","size":354,"stargazers_count":43,"open_issues_count":4,"forks_count":9,"subscribers_count":5,"default_branch":"master","last_synced_at":"2025-04-03T20:23:18.971Z","etag":null,"topics":["knowledge-embedding"],"latest_commit_sha":null,"homepage":null,"language":"C++","has_issues":true,"has_wiki":null,"has_pages":null,"mirror_url":null,"source_name":null,"license":"mit","status":null,"scm":"git","pull_requests_enabled":true,"icon_url":"https://github.com/thunlp.png","metadata":{"files":{"readme":"README.md","changelog":null,"contributing":null,"funding":null,"license":"LICENSE","code_of_conduct":null,"threat_model":null,"audit":null,"citation":null,"codeowners":null,"security":null,"support":null,"governance":null,"roadmap":null,"authors":null,"dei":null,"publiccode":null,"codemeta":null}},"created_at":"2017-05-22T03:24:49.000Z","updated_at":"2024-03-14T13:15:50.000Z","dependencies_parsed_at":null,"dependency_job_id":"bcee0b00-9d31-448e-ac07-0a60fec0cff0","html_url":"https://github.com/thunlp/IKRL","commit_stats":null,"previous_names":[],"tags_count":0,"template":false,"template_full_name":null,"repository_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/thunlp%2FIKRL","tags_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/thunlp%2FIKRL/tags","releases_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/thunlp%2FIKRL/releases","manifests_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/thunlp%2FIKRL/manifests","owner_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners/thunlp","download_url":"https://codeload.github.com/thunlp/IKRL/tar.gz/refs/heads/master","host":{"name":"GitHub","url":"https://github.com","kind":"github","repositories_count":250808192,"owners_count":21490622,"icon_url":"https://github.com/github.png","version":null,"created_at":"2022-05-30T11:31:42.601Z","updated_at":"2022-07-04T15:15:14.044Z","host_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub","repositories_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories","repository_names_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repository_names","owners_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners"}},"keywords":["knowledge-embedding"],"created_at":"2024-11-10T18:35:57.421Z","updated_at":"2025-04-25T11:31:25.650Z","avatar_url":"https://github.com/thunlp.png","language":"C++","readme":"# IKRL\n\nImage-embodied Knowledge Representation Learning (IJCAI-2017)\n\nNew: Add dataset\n\n\n# INTRODUCTION\n\nImage-embodied Knowledge Representation Learning (IKRL)\n\nImage-embodied Knowledge Representation Learning (IJCAI-2017)\n\nWritten by Ruobing Xie\n\n\n# COMPILE \n\nJust type make in the folder ./\n\n\n# DATA\n\nWe use a new dataset WN9-IMG, with triples extracted from WN18 and images extracted from ImageNet.\n\nThere are additional files needed in training, pre-training is optional:\n\n1. image2vec_fc7.txt: image feature vector, pre-trained by AlexNet (fc7 layer)\n2. (optional) entity2vec.unif / relation2vec.unif: entity \u0026 relation vector, pre-trained by TransE\n3. (optional) image_mat.unif: image projection matrix, pre-trained by IKRL (AVG)\n\nFor image2vec_fc7.txt, it is a file that contains the image feature vectors of all images learned by AlexNet (or other models). Due to its large size, we do not release our features. However, it is very easy to use other visual features learned by updated visual encoders and images.\n\nStep1. Collect (multiple) images for all entities.\n\nStep2. Learn image feature vectors for all images by AlexNet (or other models). We use the last hidden embedding (fc7 layer) as IKRL's image features.\n\nPlease refer to the original paper for more details:\n\nRuobing Xie, Zhiyuan Liu, Huanbo Luan, Maosong Sun. Image-embodied Knowledge Representation Learning. The 26th International Joint Conference on Artificial Intelligence (IJCAI'17).\n\n\n# RUN\n\ntrain: time ./Train_transI -size 50 -margin 4 -method 0\n\ntest: ./Test unif\n\n\n# CITE\n\nIf the codes or datasets help you, please cite the following paper:\n\nRuobing Xie, Zhiyuan Liu, Huanbo Luan, Maosong Sun. Image-embodied Knowledge Representation Learning. The 26th International Joint Conference on Artificial Intelligence (IJCAI'17).\n","funding_links":[],"categories":[],"sub_categories":[],"project_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fthunlp%2Fikrl","html_url":"https://awesome.ecosyste.ms/projects/github.com%2Fthunlp%2Fikrl","lists_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fthunlp%2Fikrl/lists"}